impulse

Llama 4 Maverick

Open weights

Llama 4 Maverick 17B Instruct (128E) is a high-capacity multimodal language model from Meta, built on a mixture-of-experts (MoE) architecture with 128 experts and 17 billion active parameters per forward…

API id llama-4-maverickReleased 5 Apr 2025 (Apr 2025)
Input / 1M
$0.20
at Vercel AI Gateway
Output / 1M
$0.60
at Vercel AI Gateway
Context window
131K
tokens
Max output
8K
tokens per reply
Hosts
3
3 listings
Released
Apr 2025
5 Apr 2025

Pricing

List prices at Vercel AI Gateway, in US dollars per million tokens.

TierInput / 1MOutput / 1M
Standard$0.20$0.60

What it costs a month

At Vercel AI Gateway list prices, 30 days.

  • Chat assistant5,000 requests a day, 2K in / 0.5K out$105
  • RAG over documents2,000 requests a day, 8K in / 0.6K out$118
  • Coding agent1,000 requests a day, 40K in / 2K out$276
  • Overnight batch20,000 requests a day, 4K in / 0.4K out$624
Use your own numbers

Where to run it

3 hosts serve Llama 4 Maverick. The comparison is on blended price, 3 parts input to 1 part output.

HostIn / out per 1M
llama-4-maverick
$0.20 / $0.60
llama-4-maverick
$0.334 / $1.34
+95% vs headline
databricks-llama-4-maverick
$0.50 / $1.50
+150% vs headline

Capabilities

Core

  • Streaming
  • Function calling
  • JSON mode
  • Structured outputs

Multimodal

  • Vision
  • Image input
  • Audio
  • Image generation
  • Video

Advanced

  • Reasoning
  • Tool calling
  • MCP
  • Prompt caching
  • Batch API
  • Embeddings
Takes in
Text
Produces
Text
Open weights
License: Llama Community License
Runtimes: Ollama, TGI, llama.cpp, vLLM

Cheaper alternatives

Current models that cost less and keep most of what this one can do: tools, vision, reasoning, structured output, caching and enough context.

Same model, cheaper host

No host sells it for less than Vercel AI Gateway.

Smaller Meta models

Other frontier labs

  • Claude Haiku 5.5
    Anthropic · on Anthropic
    $0.10 / $0.50
    Has every key capability
    33% less
  • GPT-6 Luna
    OpenAI · on OpenAI
    $0.10 / $0.50
    Has every key capability
    33% less
  • GPT-4o Mini
    OpenAI · on OpenAI
    $0.15 / $0.60
    Has every key capability
    13% less

Open-weight picks

  • Mistral Small 2603
    Mistral AI · on Mistral AI
    $0.15 / $0.60
    Has every key capability
    13% less
  • Qwen3.8 27B
    Alibaba Qwen · on EmpirioLabs AI
    $0.17 / $0.50
    Has every key capability
    16% less
  • Nemotron 3.5 Super VL
    NVIDIA · on Kyma API
    $0.081 / $0.472
    Has every key capability
    40% less
  • Gemma 4 31B IT
    Google · on W&B Inference
    $0.10 / $0.34
    Has every key capability
    47% less

Model line

Newer and older versions, at each lab's own price.

  1. Apr 2025Llama 4 MaverickLatest$0.20 / $0.60
  2. Apr 2025Llama4 Maverick$0.24 / $0.97

Call it

Using Vercel AI Gateway's API.

Python
from litellm import completion

response = completion(
    model="vercel_ai_gateway/llama-4-maverick",
    messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)

Quick answers

How much does Llama 4 Maverick cost?

On Vercel AI Gateway, Llama 4 Maverick costs $0.20 per million input tokens and $0.60 per million output tokens.

Which APIs offer Llama 4 Maverick?

Llama 4 Maverick runs on 3 APIs: Vercel AI Gateway, Kyma API and Databricks.

What is Llama 4 Maverick's context window?

131K tokens, with up to 8K tokens in one reply.

Is Llama 4 Maverick being retired?

No retirement date is scheduled for it on any API we track.

Is there a newer version of Llama 4 Maverick?

No. Llama 4 Maverick is the newest model in its line.

What are cheaper alternatives to Llama 4 Maverick?

Models that cost less and keep most of what it can do: Muse Spark 1.3 Contributor on Meta Model API (58% less), Claude Haiku 5.5 on Anthropic (33% less), GPT-6 Luna on OpenAI (33% less) and Mistral Small 2603 on Mistral AI (13% less). Savings are on a mix of 3 input tokens to 1 output token.

When was Llama 4 Maverick released?

Meta released it on 5 Apr 2025.

What is Llama 4 Maverick's API model id?

Use llama-4-maverick with Vercel AI Gateway. Other hosts use their own ids: databricks-llama-4-maverick on Databricks.

Sources

Raw catalog entry for llama-4-maverick on Vercel AI Gateway
{
  "input_cost_per_token": 2e-7,
  "litellm_provider": "vercel_ai_gateway",
  "max_input_tokens": 131072,
  "max_output_tokens": 8192,
  "mode": "chat",
  "output_cost_per_token": 6e-7,
  "supports_tool_choice": true
}